Metraipolis › archive › dc_584d2d6fe4f2
Thompson Sampling for Learning in Zero-Sum Games — arXiv:2610.18899
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.18899
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T14:12:25.960Z
- SHA-256 of the archived text
6400d9dd5b78078a63c5b5fbdf48686c57130e6bb09bd2ea96ea4dcde6329c8f- Stable URL of this page
https://metraipolis.com/documento/dc_584d2d6fe4f2
How to cite
Thompson Sampling for Learning in Zero-Sum Games — arXiv:2610.18899 — arXiv (en), https://arxiv.org/abs/2610.18899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_584d2d6fe4f2 (Metraipolis archive dc_584d2d6fe4f2, sha256 6400d9dd5b78078a63c5b5fbdf48686c57130e6bb09bd2ea96ea4dcde6329c8f)
BibTeX
@misc{dc584d2d6fe4f2,
title = {Thompson Sampling for Learning in Zero-Sum Games — arXiv:2610.18899},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.18899},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_584d2d6fe4f2; sha256 6400d9dd5b78078a63c5b5fbdf48686c57130e6bb09bd2ea96ea4dcde6329c8f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.18899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Thompson sampling (TS) achieves logarithmic and asymptotically optimal regret in classical stochastic bandits, but its behavior in multi-agent learning remains less understood. We study TS in an unknown finite two-player zero-sum matrix game with Gaussian bandit feedback. Under centralized self-play, TS samples a single payoff matrix from the posterior and plays a Nash equilibrium of the sampled game. For every fixed zero-sum game with a unique Nash equilibrium, we show that the expected cumulative duality gap is $O(\sqrt T)$. The rate improves to $O(\log T)$ for pure equilibria. We then identify the exact coefficient of the $\sqrt T$ term, which depends only on the equilibrium strategies and the noise level. We also establish a minimax lower bound matching the square-root dependence on $T$ and an instance-dependent lower bound matching the logarithmic dependence on $T$ for a class of games with a unique pure equilibrium. Experiments support the predicted rates and coefficient. Tema degli esploratori: economia dei servizi IA (elenco arXiv cs.GT, nuovi lavori del giorno). Categorie arXiv: Computer Science and Game Theory (cs.GT) ; Machine Learning (stat.ML). Pagina del lavoro: https://arxiv.org/abs/2610.18899 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T13:52:05.389Z dalla pagina https://arxiv.org/list/cs.GT/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).